An order sorting and path optimization method and system based on real-time logistics data

Abstract

The application provides an order sorting and path optimization method and system based on real-time logistics data, and relates to the technical field of logistics task adjustment and dispatching. The method comprises the following steps: acquiring real-time traffic data of a city road network, and determining the congestion level of a road section or a region based on the real-time traffic data; monitoring a distribution task in real time, and acquiring the position of a delivery person, a task order list, an order state and an originally planned route; when the congestion level of a road section or a region on the current or planned route of the delivery person changes, determining an adjustment strategy by consulting a preset congestion level classification response rule library; and sending new route, order adjustment information or re-dispatching information to the delivery person terminal in real time. The method can cope with sudden traffic changes, guarantee the smooth completion of a distribution task and the timeliness of an order, and is beneficial to responding to the demand for urban instant distribution.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of logistics task adjustment and distribution, in particular to an order sorting and path optimization method and system based on real-time logistics data. BACKGROUND

[0002] The urban instant delivery platform is an important part of the modern urban logistics system, aiming to meet the needs of users for fast delivery of goods. During the operation of the platform, the system continuously receives a large number of orders from different areas and different merchants in the city. In order to fulfill the delivery time commitment to users, the platform needs to efficiently process these incoming orders. This usually includes preliminary sorting of orders, such as integrating multiple orders from the same merchant or similar geographic locations to form batch tasks, or preliminary allocation based on the current location, delivery area and load of the delivery personnel. After sorting and distribution are completed, the system generates a delivery task for each delivery personnel containing one or more orders, and based on the current traffic conditions and geographic information, plans the optimal transportation route from the current location of the delivery personnel to each pickup point (merchant) and finally to each delivery point (user). The delivery personnel executes the task according to the route planned by the system until all allocated orders are delivered.

[0003] However, the urban traffic environment has a high degree of dynamicity and unpredictability. Unplanned incidents may occur at any time on the road, such as traffic accidents, vehicle breakdowns, temporary road construction, traffic control, or simply regional or road congestion due to a surge in vehicle flow. These changes in traffic conditions are not known in advance, but occur and evolve rapidly in real time. The instant delivery platform needs to be able to obtain dynamic data reflecting the current urban traffic conditions in real time, such as by interfacing with map service providers, traffic management departments or other data sources to obtain real-time congestion indices, average speeds, reports of specific incidents (such as accidents, road closures), etc.

[0004] When the platform receives these real-time traffic data, the delivery tasks and routes originally planned based on older traffic information may quickly become unsuitable. If the real-time traffic data shows that there is a serious traffic jam or other unfavorable conditions on the transportation route currently being traveled or about to be traveled by the delivery personnel, the originally planned path may no longer be the optimal choice, resulting in a significant increase in the travel time of the delivery personnel, which directly affects the estimated delivery time of all orders in the current delivery task. For instant delivery orders with strict time constraints, any delay in delivery can seriously affect user experience, leading to user dissatisfaction, and even cause the platform to fail to fulfill its time commitment to users, resulting in economic losses or damage to brand reputation.

[0005] Traditional delivery systems can rely on static route planning or manual experience for adjustment, and it is difficult to quickly and accurately respond to sudden and dynamically changing traffic conditions. Simply re-planning the path of the current order often fails to fully consider other undelivered orders that the delivery person may have in hand, and the adjustment of a route can have a chain effect on the estimated delivery time of subsequent orders. In addition, changes in traffic conditions not only affect the transportation route, but also can affect the sorting and allocation decisions of orders that have not yet been taken by the delivery person. For example, if the traffic in a certain area suddenly becomes extremely congested, orders that were originally allocated to delivery personnel who are active in or will soon enter the congested area may cause serious delivery delays if they continue to be delivered. At this time, the system may need to re-evaluate the allocation scheme for these orders, consider re-sorting and allocating them to delivery personnel located in other areas with relatively good traffic conditions and the ability, or adjust the order pickup and delivery sequence to prioritize orders less affected by traffic.

[0006] Therefore, how to quickly and accurately judge the impact of real-time, sudden and dynamically changing traffic data on the current order sorting state and delivery personnel transportation route in the urban instant delivery environment, and be able to generate a comprehensive response plan including dynamic adjustment of transportation route and re-sorting / allocating orders, is a complex and challenging technical problem. The entire decision and adjustment process must be completed in a very short time in order to synchronize new instructions to affected delivery personnel in time, maximize the negative impact of traffic changes, and maximize overall delivery efficiency while ensuring delivery timeliness. This requires the system to have efficient data processing, fast algorithm calculation, and real-time communication capabilities with delivery personnel terminal devices.

[0007] Currently, there is no effective technical solution to the above problems. SUMMARY

[0008] The purpose of the present application is to provide an order sorting and path optimization method and system based on real-time logistics data, which can respond to sudden traffic changes and ensure the smooth completion of delivery tasks and order timeliness, and is beneficial to respond to the needs of urban instant delivery.

[0009] In a first aspect, the present application provides an order sorting and path optimization method based on real-time logistics data, comprising the following steps:

[0010] Obtain real-time traffic data of the urban road network, and determine the congestion level of the road section or area based on the real-time traffic data;

[0011] Real-time monitoring of delivery tasks, obtaining the position of the delivery personnel, the list of task orders, the order status and the original planned route;

[0012] When the congestion level of the road section or area on the current or planned route of the delivery personnel changes, the adjustment strategy is determined by consulting the preset congestion level grading response rule library, and when the adjustment strategy is route re-planning, the optimal route for the delivery personnel is re-determined according to the real-time traffic data, and when the adjustment strategy is order re-assignment evaluation, the feasibility of re-assignment of the affected delivery personnel's task of the order not yet taken is evaluated, and it is determined whether there is a substitute delivery personnel who can complete the order on time.

[0013] The new route, order adjustment information or re-assignment notification is sent to the delivery personnel terminal in real time.

[0014] The order sorting and path optimization method based on real-time logistics data provided by the application can quickly and cooperatively judge the influence of sudden and dynamically changing real-time traffic conditions on the existing order sorting state and the delivery personnel transportation route, and generate a graded and dynamic adjustment scheme (including route re-planning and order re-assignment) to minimize delivery delays, ensure order timeliness and improve overall efficiency.

[0015] In a second aspect, the application provides an order sorting and path optimization system based on real-time logistics data, comprising:

[0016] The first acquisition module is configured to acquire real-time traffic data of a city road network, and determine the congestion level of a road section or area based on the real-time traffic data.

[0017] The second acquisition module is configured to monitor the delivery task in real time, and acquire the position of the delivery personnel, the order list of the task, the order state and the original planned route.

[0018] The strategy module is configured to determine the adjustment strategy by consulting the preset congestion level grading response rule library when the congestion level of the road section or area on the current or planned route of the delivery personnel changes, and when the adjustment strategy is route re-planning, the optimal route for the delivery personnel is re-determined according to the real-time traffic data, and when the adjustment strategy is order re-assignment evaluation, the feasibility of re-assignment of the affected delivery personnel's task of the order not yet taken is evaluated, and it is determined whether there is a substitute delivery personnel who can complete the order on time.

[0019] The sending module is configured to send the new route, order adjustment information or re-assignment notification to the delivery personnel terminal in real time.

[0020] From the above, the order sorting and path optimization method based on real-time logistics data provided by the application can obtain and analyze urban traffic data in real time, convert it into a quantitative congestion level, and quickly and accurately determine the impact of sudden traffic changes on the instant delivery task being performed. By constructing a hierarchical response rule library based on the congestion level, the system can automatically trigger differentiated and targeted adjustment strategies according to the degree of influence, avoiding excessive or insufficient intervention. The scheme cooperatively considers dynamic re-planning of the transportation route and re-dispatching of orders that have not been picked up, forming a comprehensive response mechanism. This enables the platform to efficiently respond to the dynamic urban traffic environment, timely avoid congestion areas, reasonably allocate delivery resources, and maximize the reduction of delivery delays caused by traffic changes, significantly improving the on-time delivery rate of orders and ensuring the timeliness and reliability of instant delivery services, thereby improving overall delivery efficiency and user experience.

[0021] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the application as described in the written description and claims. The objects and other advantages of the present application will be realized and attained by means of the structures particularly pointed out in the written description and claims. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 A flow chart of the order sorting and path optimization method based on real-time logistics data provided by the embodiment of the application.

[0023] Figure 2 A structural schematic diagram of the order sorting and path optimization system based on real-time logistics data provided by the embodiment of the application.

[0024] Label explanation:

[0025] 100, first acquisition module; 200, second acquisition module; 300, strategy module; 400, sending module. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the accompanying drawings of the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all. The components of the embodiments of the application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the application. Based on the embodiments of the application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the application.

[0027] It should be noted that similar reference numerals and letters refer to like items in the accompanying drawings, and once an item is defined in one drawing, it is not necessary to further define and explain it in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0028] With reference to the accompanying drawings, which are incorporated herein by reference, the present application provides an order sorting and path optimization method based on real-time logistics data, comprising the following steps: Figure 1

[0029] Obtaining real-time traffic data of urban road network, and determining congestion levels of road sections or areas based on the real-time traffic data;

[0030] Real-time monitoring of delivery tasks, obtaining the positions of delivery personnel, task order lists, order statuses, and original planned routes;

[0031] When the congestion levels of road sections or areas on the current or planned routes of the delivery personnel change, determining adjustment strategies by consulting a pre-set congestion level classification response rule library, and when the adjustment strategy is route re-planning, re-determining the optimal route for the delivery personnel based on real-time traffic data, and when the adjustment strategy is order re-dispatching evaluation, evaluating the re-dispatching feasibility of the orders not yet picked up in the tasks of the affected delivery personnel, and determining whether there is a substitute delivery personnel who can complete the orders on time;

[0032] Real-time sending of new route, order adjustment information, or re-dispatching notifications to the delivery personnel terminal.

[0033] Obtaining real-time traffic data of urban road network and determining congestion levels therefrom can be achieved by connecting an external traffic data interface, which provides speed, flow, or event information of road sections. The determination of congestion levels is based on the comparison of real-time data with pre-set thresholds. Real-time monitoring of delivery tasks involves maintaining a task database recording delivery personnel, orders, statuses, and original route information, and the positions of delivery personnel are obtained through terminal GPS data. When a change in congestion level is detected, the system consults a pre-set rule library, which determines whether to perform route re-planning or order re-dispatching evaluation according to the type and location of the congestion change. If the strategy is route re-planning, a route calculation algorithm is called, which calculates multiple alternative paths using the current real-time traffic data and selects the path with the shortest predicted travel time as the optimal route. If the strategy is order re-dispatching evaluation, the system evaluates the orders not yet picked up in the tasks of the affected delivery personnel and analyzes potential substitute delivery personnel, taking into account their positions, loads, and predicted delivery times, to determine whether there is a substitute who can complete the orders on time. Finally, new route instructions, order adjustment information, or re-dispatching notifications are sent to the terminal devices of the delivery personnel through a communication module.

[0034] ​Specifically, the method grasps the dynamic changes of traffic conditions by continuously acquiring urban traffic data and determining congestion levels. At the same time, the system tracks the execution status of delivery tasks in real time, including the positions of delivery personnel and the original planned routes. When detecting a change in the congestion level of a road segment or area on the current or planned route of a delivery personnel, the system does not perform fixed operations, but determines a coping strategy according to a pre-set rule base. If the strategy is route re-planning, the system uses the latest real-time traffic data to calculate and determine a new optimal route for the delivery personnel to avoid congested areas and reduce delivery time. If the strategy is order re-assignment evaluation, the system evaluates the orders that have not been picked up in the delivery personnel's tasks affected by traffic, determines whether they can be reassigned to other delivery personnel, and determines whether there are alternative delivery personnel who can complete these orders on time. In this way, the method can dynamically cope with the impact of real-time traffic changes on delivery tasks, maintain delivery efficiency and timeliness through route adjustment or order re-assignment. New route, order adjustment information or re-assignment notification is sent to the delivery personnel terminal in real time to guide the delivery personnel to execute the new task arrangement.

[0035] In some embodiments, for example, a delivery personnel is executing a task containing multiple orders, and the original planned route passes through a major road segment in the city center. The real-time traffic data monitoring system detects that the congestion level of the road segment changes rapidly from "smooth" to "congestion", and determines that the congestion level has changed. The system consults the pre-set congestion level classification response rule base, and according to the congestion change, determines that the adjustment strategy is "route re-planning". The system immediately recalculates multiple alternative routes from the current position of the delivery personnel to the next destination according to the current real-time traffic data. One of the alternative routes bypasses the congested road segment, although the distance increases by 800 meters, but according to the real-time traffic data prediction, the expected travel time is reduced by 15 minutes compared to the original route. The system selects the route with the shortest expected travel time as the new optimal route. The new route instructions, including detailed navigation information, are sent to the terminal device of the delivery personnel in real time to guide him to continue the delivery task along the new route.

[0036] In certain embodiments, when the adjustment strategy is route re-planning, the step of recalculating the optimal route for the delivery personnel according to the real-time traffic data includes:

[0037] Obtaining historical delivery data of the delivery personnel, and constructing a personalized route preference library of the delivery personnel by extracting route segments frequently selected by the delivery personnel;

[0038] When re-planning the route, multiple alternative routes are calculated according to the real-time traffic data, and combined with the personalized route preference library of the delivery personnel, the alternative routes are weighted and scored according to the degree of coincidence between the preferred route of the delivery personnel and the alternative routes; wherein the higher the degree of coincidence with the preferred route of the delivery personnel, the higher the score;

[0039] selecting the route with the highest score as the optimal route.

[0040] The step of obtaining historical delivery data of the delivery personnel and constructing a personalized route preference library of the delivery personnel by extracting route segments frequently selected by the delivery personnel can be implemented by analyzing the trajectory data of the delivery tasks completed by the delivery personnel in the past. The system records the actual driving path of the delivery personnel for each delivery task, and decomposes these paths into route segments. Through statistical analysis, the combination of route segments or path segments with high frequency of use by the delivery personnel between similar starting points and ending points (or passing points) is identified. These high-frequency path segments are stored and constitute the personalized route preference library of the delivery personnel. In the step of calculating a plurality of candidate routes according to real-time traffic data, a standard path planning algorithm such as the Dijkstra algorithm can be used to search on the urban road network model in combination with real-time road travel times (calculated according to real-time traffic data) to generate a plurality of feasible paths from the current location to the next destination (pick-up point or delivery point). In the step of weighting and scoring the candidate routes according to the degree of coincidence between the preferred route of the delivery personnel and the candidate routes, the proportion of common route segments or the similarity index of each candidate route and the relevant path segments in the preference library can be calculated. For example, the proportion of the length of the intersection route segments in the total length of the candidate route can be calculated, or a graph matching algorithm can be used to calculate the similarity. The calculated degree of coincidence is used as a weight or score factor for the evaluation of the candidate routes. The higher the degree of coincidence between the preferred route of the delivery personnel and the candidate route, the higher the weighted score obtained by the candidate route. In the step of selecting the route with the highest score as the optimal route, the weighted scores of all generated candidate routes are compared, and the route with the highest score value is selected as the optimal route recommended to the delivery personnel.

[0041] Specifically, when the adjustment strategy is route re-planning, the method first obtains the historical delivery trajectory data of the delivery person. The system processes these historical trajectories, identifies the specific road segments or path combinations that the delivery person tends to choose in different areas or different task types, and builds a database reflecting his individual route preferences. When real-time traffic conditions change trigger route re-planning, the system uses the current real-time traffic data (e.g., real-time congestion index or estimated travel time of each road segment) to calculate multiple feasible alternative routes from the delivery person's current location to his next destination (e.g., the next pickup point). These alternative routes are possible optimal paths calculated based on real-time traffic conditions. Then, the system compares these alternative routes with the delivery person's personalized route preference library. Calculate the degree of coincidence of each alternative route with the relevant preference path in the preference library, for example, calculate the proportion of the number or length of road segments they share. Based on the degree of coincidence, each alternative route is given a weighted score. The higher the degree of coincidence, the higher the score, which reflects that the route is closer to the delivery person's habitual path. Finally, the system selects the alternative route with the highest weighted score as the new optimal route recommended to the delivery person. This method aims to plan a route that not only effectively avoids congestion, but also maintains high consistency with the delivery person's familiar or habitual path by considering real-time traffic conditions and individual route preferences, thereby improving the delivery person's acceptance and execution efficiency of the new route, reducing delays that may be caused by unfamiliarity with the new route, and improving overall delivery efficiency.

[0042] In some embodiments, when the adjustment strategy is route re-planning, the step of re-determining the optimal route for the delivery person according to real-time traffic data includes:

[0043] Obtain the historical delivery data of the delivery person, calculate the average delivery time of the delivery person in different time periods and different areas, and build a personalized delivery time model for the delivery person;

[0044] When the adjustment strategy is route re-planning, calculate multiple alternative routes according to real-time traffic data, combine the delivery person's personalized delivery time model, and predict the estimated delivery time of the delivery person on each alternative route; the estimated delivery time is determined by the route length, real-time traffic conditions and the delivery person's personalized delivery time model;

[0045] Select the route with the shortest estimated delivery time as the optimal route.

[0046] The historical delivery data of the delivery personnel is obtained. By analyzing the trajectory data of the delivery personnel completing historical delivery tasks, order completion timestamps and other information, the average driving speed of the delivery personnel in a specific time period (for example, peak hours on weekdays, off-peak hours on weekends) and a specific geographic area (for example, urban central area, suburb) or the average time consumption per unit distance / order can be calculated. Thus, a personalized delivery time model reflecting the individual delivery efficiency difference of the delivery personnel is constructed. The model can be stored as a lookup table or a parameterized function form. When route replanning is needed, the system calculates multiple feasible alternative routes from the current location of the delivery personnel to the next destination (pick-up point or delivery point) according to the current real-time traffic data, for example, by calling a map service API to obtain current traffic information. For each alternative route, the system decomposes it into multiple road segments and combines the current real-time traffic conditions of the road segment (for example, congestion level, average vehicle speed) and the efficiency parameters of the corresponding time period and area in the delivery personnel's personalized delivery time model to predict the time required to pass through the road segment. The predicted times of all road segments are added up to obtain the predicted total arrival time of the alternative route.

[0047] Specifically, the technical solution aims to solve the problem of how to select the most suitable route for a specific delivery personnel in terms of individual efficiency when real-time traffic congestion requires route replanning. First, the system continuously collects and analyzes the historical delivery data of each delivery personnel to establish a personalized model reflecting their delivery efficiency under different conditions. When traffic congestion changes trigger route replanning, the system quickly calculates multiple feasible alternative routes using real-time urban traffic data. Subsequently, for each alternative route, the system no longer relies solely on general traffic models to predict time, but incorporates the delivery personnel's personalized delivery time model into the calculation process. This means that the system takes into account the delivery personnel's historical performance under similar road conditions, time periods, and areas, thus more accurately predicting the actual time required to complete the route. For example, if the personalized model shows that a delivery personnel's actual driving speed in a specific area is usually higher than the average, the system will make corresponding adjustments when predicting the time required to pass through the road segment in that area. In this way, the system can generate a predicted arrival time for each alternative route that is more in line with the actual situation of the delivery personnel. Finally, the system compares the predicted arrival times of all alternative routes and selects the shortest one as the optimal route recommended to the delivery personnel. In this way, the solution can overcome the limitations of relying solely on general traffic data for route optimization, improve the accuracy of predicted arrival times, and thus enable the selection of routes that are more optimal for individual delivery personnel and more likely to be completed on time, improving overall delivery efficiency and timeliness.

[0048] In some embodiments, when the adjustment strategy is route replanning, the step of re-determining the optimal route for the delivery personnel according to real-time traffic data includes:

[0049] obtaining the remaining power information of the delivery person terminal;

[0050] When the adjustment strategy is route re-planning, multiple alternative routes are calculated according to real-time traffic data, and the power required for the delivery person to reach all subsequent points to be delivered according to each alternative route is predicted (by default, using navigation throughout the journey) in combination with the remaining power information of the delivery person terminal;

[0051] Based on the predicted power and the remaining power of the delivery person terminal, the optimal route for the delivery person is re-determined according to real-time traffic data.

[0052] Obtaining the remaining power information of the delivery person terminal Through the delivery application running on the delivery person terminal device, the battery percentage or remaining available time information of the device is collected in real time, and the information is uploaded to the platform system. According to real-time traffic data, multiple alternative routes are calculated Using real-time traffic data provided by map service providers, combined with urban road network data, path search algorithms are used to calculate multiple feasible paths from the current location of the delivery person to all subsequent points to be delivered in their tasks. These paths take into account the current traffic conditions to estimate travel time. In combination with the remaining power information of the delivery person terminal, the power required for the delivery person to reach all subsequent points to be delivered according to each alternative route is predicted For each calculated alternative route, the system predicts the total power required to complete the route according to the length of the route, the estimated driving time affected by real-time traffic, and the typical power consumption model when the terminal device uses the navigation function. The prediction model can be established based on historical data or device specifications, and the default is to estimate the entire journey with navigation turned on. Based on the predicted power and the remaining power of the delivery person terminal, the optimal route for the delivery person is re-determined Compare the predicted power consumption of each alternative route with the current remaining power of the delivery person terminal. Eliminate those routes whose predicted power consumption is greater than or equal to the remaining power. Among the remaining feasible routes, the final optimal route is determined in combination with real-time traffic data. This process ensures that the selected route is feasible under the current traffic conditions, and the delivery person terminal has enough power to complete the task.

[0053] In particular, the technical solution aims to solve the problem that the power of the delivery terminal is not considered when the delivery route is replanned according to real-time traffic data, which may cause the route to be unable to be executed smoothly. When the system determines that the congestion level of the current or planned route of the delivery personnel changes and needs to be replanned, the current remaining power information of the delivery terminal is first obtained. At the same time, the system calculates a plurality of possible alternative routes according to the current real-time traffic data. For each alternative route, the system will combine the remaining power information of the delivery terminal to predict the power required by the delivery personnel to complete the remaining delivery task according to the route, and the prediction process will consider the use of navigation by default. This prediction quantifies the power consumption of each route. Finally, the system comprehensively considers the predicted power consumption and the actual remaining power of the delivery terminal. Based on these two pieces of information, and combined with real-time traffic data, the system selects an optimal route from the alternative routes. This selection process ensures that the selected route is not only efficient in traffic conditions, but also feasible in terms of power, avoiding delivery interruption due to insufficient power. By obtaining the remaining power information, power constraints are provided. By calculating alternative routes, multiple path options are provided. By predicting the required power, the power cost of each route is evaluated. Determining the optimal route based on the predicted power and the remaining power of the delivery terminal integrates power constraints into the route selection process and ensures the feasibility of the final route execution.

[0054] In some embodiments, based on the predicted power and the remaining power of the delivery terminal, the step of re-determining the optimal route for the delivery personnel according to real-time traffic data comprises:

[0055] Obtain historical delivery data of the delivery personnel, extract frequently selected route segments of the delivery personnel, and construct a personalized route preference library of the delivery personnel;

[0056] According to the personalized route preference library of the delivery personnel, the familiarity of the delivery personnel with each alternative route is judged (the familiarity can be quantified by a score), and the following steps A1-A4 are performed:

[0057] A1. If the familiarity of the delivery personnel with the alternative route is less than a preset familiarity threshold (i.e., the alternative route is an unfamiliar route), and the predicted power is greater than or equal to the remaining power of the delivery terminal, the route is excluded;

[0058] A2. If the familiarity of the delivery personnel with the alternative route is less than the familiarity threshold, and the predicted power is less than the remaining power of the delivery terminal, the route is retained;

[0059] A3. If the familiarity of the deliveryman to the alternative route is greater than or equal to the familiarity threshold (i.e. the alternative route is a familiar route), and the predicted power is greater than or equal to the remaining power of the deliveryman terminal, then the predicted power is adjusted downward, and if the adjusted power is greater than or equal to the remaining power of the deliveryman terminal, then the route is eliminated; if the adjusted power is less than the remaining power of the deliveryman terminal, then the route is retained (the deliveryman is able to complete the delivery task without using navigation all the way, so the predicted power based on navigation all the way is too extreme, and there is actually room for adjustment);

[0060] A4. If the familiarity of the deliveryman to the alternative route is greater than or equal to the familiarity threshold, and the predicted power is less than the remaining power of the deliveryman terminal, then the route is retained;

[0061] Based on the remaining alternative routes, the optimal route for the deliveryman is re-determined according to real-time traffic data.

[0062] In order to solve the problem that the familiarity of the deliveryman to the route is not considered when determining the optimal route according to the predicted power and the remaining power, the historical delivery data of the deliveryman is obtained, and the route segments frequently selected by the deliveryman are extracted therefrom to establish a personalized route preference library of the deliveryman. The preference library is used to quantify the familiarity of the deliveryman to different alternative routes. When the route is re-planned, the system calculates a plurality of alternative routes according to real-time traffic data. For each alternative route, the system judges the familiarity of the deliveryman to the route in combination with the personalized route preference library of the deliveryman. At the same time, the system predicts the power required by the deliveryman to reach all subsequent to-be-delivered points according to the alternative route. Then, the system compares the familiarity of the deliveryman to the alternative route with a preset familiarity threshold, and compares the predicted power with the remaining power of the deliveryman terminal, and screens the alternative route according to a preset logic (steps A1-A4). If the alternative route is unfamiliar and the predicted power is greater than or equal to the remaining power, it is considered that the route has a power risk, and it is eliminated. If the alternative route is unfamiliar but the predicted power is less than the remaining power, it is retained. If the alternative route is familiar and the predicted power is greater than or equal to the remaining power, it is considered that the deliveryman can not need to rely on navigation all the way, and the system adjusts the predicted power downward, and if the power is still insufficient after adjustment, the route is eliminated, otherwise it is retained. If the alternative route is familiar and the predicted power is less than the remaining power, it is retained. Through the above screening process, the routes with insufficient power or high risk are eliminated. Finally, among the alternative routes retained after screening, the system selects a route as the optimal route in combination with real-time traffic data. Thus, by introducing the route familiarity information of the deliveryman, the deficiency of predicting the power based on navigation all the way is corrected, the route selection is more in line with the actual situation, the selection of the route in the case of insufficient power and unfamiliarity is avoided, and more practical routes are retained when the route is familiar.

[0063] In some embodiments, assume the delivery agent terminal has 30% battery left. The system calculates three candidate routes R1, R2, R3. By analyzing the delivery agent’s personalized route preference library, the system judges that the delivery agent’s familiarity score for R1 is 2 (below the familiarity threshold of 5, determining that it is an unfamiliar route), the familiarity score for R2 is 8 (above the familiarity threshold of 5, determining that it is a familiar route), and the familiarity score for R3 is 7 (above the familiarity threshold of 5, determining that it is a familiar route). The system predicts that the delivery agent will need 35% battery to complete the task along R1, 32% battery along R2, and 28% battery along R3. According to the screening logic: for R1, the familiarity is less than the threshold (unfamiliar), and the predicted battery needed is 35%, which is greater than the remaining battery of 30%, according to Al, R1 is eliminated. For R2, the familiarity is greater than or equal to the threshold (familiar), and the predicted battery needed is 32%, which is greater than the remaining battery of 30%, according to A3, the predicted battery is adjusted, for example, to 29%. The adjusted battery of 29% is less than the remaining battery of 30%, according to A3, R2 is retained. For R3, the familiarity is greater than or equal to the threshold (familiar), and the predicted battery needed is 28%, which is less than the remaining battery of 30%, according to A4, R3 is retained. After screening, the remaining candidate routes are R2 and R3. The system selects the optimal route from R2 and R3 according to real-time traffic data.

[0064] In certain embodiments, when the adjustment strategy is order reassignment evaluation, the feasibility of reassigning the orders in the affected delivery agent’s task that have not been picked up is evaluated, and it is determined whether there is an alternative delivery agent who can complete the order on time:

[0065] Key information of the orders in the affected delivery agent’s task that have not been picked up is extracted; the key information includes the estimated delivery time, the order value, and the distance information between each alternative delivery agent and the pickup location;

[0066] According to the key information, the reassignment scores of each alternative delivery agent are calculated; the shorter the estimated delivery time, the higher the order value, and the closer the distance, the higher the score;

[0067] Alternative delivery agents with a reassignment score higher than a preset score threshold are screened, and if there are any, it is determined that there is an alternative delivery agent who can complete the order on time.

[0068] The key information of the orders not yet picked up in the affected delivery person's tasks is extracted, including the estimated delivery time, the order value, and the distance information of each replacement delivery person to the pickup location. The estimated delivery time reflects the timeliness requirement of the order, the order value can affect the priority of the order, and the distance is directly related to whether the replacement delivery person can quickly reach the pickup location. According to the key information, the reassignment score of each replacement delivery person is calculated. The scoring rule is set as: the shorter the estimated delivery time, the higher the order value, and the closer the distance between the replacement delivery person and the pickup location, the higher the score. By considering these factors comprehensively and quantifying them as scores, the system can objectively compare different replacement delivery persons. Replacement delivery persons with a reassignment score higher than a preset score threshold are screened. If such a delivery person exists, it is determined that there is a replacement delivery person who can complete the order on time. This method based on key information extraction, multi-factor score calculation, and threshold screening provides a systematic and quantifiable process for evaluating the suitability of replacement delivery persons.

[0069] Specifically, when it is determined that the congestion level of a road segment or area on the current or planned route of a delivery person changes, and the adjustment strategy is order reassignment evaluation, the system first identifies the affected delivery person and the orders not yet picked up in his / her tasks. For these affected orders not yet picked up, the system extracts their key information, such as the latest delivery time required by the order (from which the estimated delivery time is calculated), the value of the goods included in the order, and the pickup location of the order. At the same time, the system obtains the location information of all potential replacement delivery persons in the current area and calculates their respective distances to the pickup location of the order. Based on these extracted key information, the system evaluates each potential replacement delivery person and calculates a reassignment score. This score takes into account the urgency of the order (reflected by the estimated delivery time), the importance of the order (reflected by the order value), and the response speed of the replacement delivery person (reflected by the distance to the pickup location). The closer the estimated delivery time (i.e., the higher the timeliness requirement), the higher the order value, and the closer the distance between the replacement delivery person and the pickup location, the higher the calculated score, indicating that the replacement delivery person is more suitable to take over the order. After the calculation is completed, the system compares the scores of all potential replacement delivery persons with a preset score threshold. Only replacement delivery persons with a score higher than the threshold are considered as candidates with the potential to complete the order on time. If there is at least one replacement delivery person with a score higher than the threshold after screening, the system determines that the affected order has the feasibility of being reassigned to other delivery persons and completed on time. In this way, the system can quickly identify suitable replacement solutions, avoid resource waste or new delivery delays caused by blind allocation, and ensure overall delivery efficiency.

[0070] In some embodiments, the step of calculating the reassignment score of each replacement delivery person according to the key information includes:

[0071] According to the key information, the preset order type library is queried to determine the required delivery skills or equipment of the order; the order type library stores the required delivery skills or equipment information of different order types;

[0072] According to the historical delivery data of each alternative delivery person, whether each alternative delivery person has the required delivery skills or equipment of the order is analyzed, and if not, the reassignment score of the alternative delivery person is set to zero.

[0073] According to the key information of the order, the system queries the preset order type library to obtain the specific skill or equipment requirement required for completing the delivery of the order. The order type library is constructed as a data structure that stores the correspondence between different order types and required delivery capabilities. Further, the system analyzes the historical delivery data of each potential alternative delivery person, which records the order types handled by the delivery person in the past, the equipment used or the qualifications obtained. Thus, the system determines whether the alternative delivery person has the skills or equipment specified for the order in the order type library. If the determination result is not, the reassignment score of the alternative delivery person for this order is set to zero. Through this process, only the delivery person who has the required ability to complete a specific order is included in the subsequent scoring and screening process, improving the accuracy of the reassignment decision and supporting the timely completion of orders.

[0074] Specifically, when it is necessary to evaluate the feasibility of reassigning an order that has not yet been picked up by a delivery person affected by traffic congestion, the system first extracts the key information of the order to be reassigned, such as the order type. Based on this key information, the system queries the internally stored order type library, which predefines the requirements of different order types (such as fresh food, large items, medicine, etc.) on the delivery person's ability (such as the need for a cooler, the need for a van, the need for a specific qualification, etc.). Thus, the required delivery skills or equipment of the current order are determined. Subsequently, the system obtains the historical delivery data of all potential alternative delivery persons and analyzes these data to determine whether each alternative delivery person meets the skill or equipment requirements of the order. For example, by analyzing historical order records or delivery person profile information, it is determined whether the delivery person has ever delivered an order requiring a cooler or whether he or she has registered a van. If the analysis result shows that a certain alternative delivery person does not have the skills or equipment required to complete the order, regardless of his or her performance in other evaluation dimensions (such as distance, estimated delivery time), the reassignment score of the alternative delivery person for this order will be directly set to zero. This effectively excludes delivery persons who do not have the corresponding ability from the candidate list for reassignment, thereby avoiding the assignment of orders to delivery persons who cannot complete the task, improving the success rate of order reassignment and delivery efficiency.

[0075] In some embodiments, consider a "frozen food" order that needs to be reassigned. The system extracts the order key information and identifies the order type as "frozen food". The system queries the order type library, which stores the rule that "frozen food" orders require "insulated boxes". The system obtains the historical delivery data of the substitute delivery person A and the substitute delivery person B. The historical data of delivery person A shows that he often delivers fresh orders, and his profile information indicates that he is equipped with an insulated box. The historical data of delivery person B shows that he mainly delivers files and small items, and his profile information does not indicate that he is equipped with an insulated box. The system analyzes the data of delivery person A and determines that he has "insulated box" and meets the order requirements, and his reassignment score will be calculated based on other factors. The system analyzes the data of delivery person B and determines that he does not have "insulated box" and does not meet the order requirements, and his reassignment score for this "frozen food" order is set to zero. Therefore, delivery person B is excluded from the reassignment candidate for this "frozen food" order, ensuring that the order is assigned to a delivery person with insulation capability, avoiding food spoilage, and ensuring delivery quality.

[0076] Reference is made to the accompanying drawings Figure 2 The present application provides an order sorting and path optimization system based on real-time logistics data, comprising:

[0077] The first acquisition module 100 is used to acquire real-time traffic data of urban road network, and determine the congestion level of road section or area based on the real-time traffic data;

[0078] The second acquisition module 200 is used to monitor the delivery task in real time, acquire the position of delivery person, task order list, order status and original planned route;

[0079] The strategy module 300 is used to determine the adjustment strategy by consulting the pre-set congestion level classification response rule library when the congestion level of road section or area on the current or planned route of the delivery person changes, and re-determine the optimal route for the delivery person according to the real-time traffic data when the adjustment strategy is route re-planning, and evaluate the reassignment feasibility of the orders not yet picked up in the affected delivery person's task when the adjustment strategy is order reassignment evaluation, and determine whether there is a substitute delivery person who can complete the order on time;

[0080] The sending module 400 is used to send the new route, order adjustment information or reassignment notification to the delivery person terminal in real time.

[0081] In some embodiments, the strategy module 300 is used to execute when re-determining the optimal route for the delivery person according to the real-time traffic data when the adjustment strategy is route re-planning:

[0082] Obtain the historical delivery data of the delivery person, and construct the personalized route preference library of the delivery person by extracting the route segments often selected by the delivery person;

[0083] In the route re-planning, a plurality of candidate routes are calculated according to the real-time traffic data, and the candidate routes are weighted and scored according to the coincidence degree of the delivery person's preferred route and the candidate route in combination with the delivery person's personalized route preference library; the higher the coincidence degree of the route with the delivery person's preferred route, the higher the score;

[0084] The route with the highest score is selected as the optimal route.

[0085] In some embodiments, the strategy module 300 is used to perform the following when the adjustment strategy is route re-planning and the optimal route for the delivery person is re-determined according to the real-time traffic data:

[0086] The historical delivery data of the delivery person is obtained, the average delivery time of the delivery person in different time periods and different areas is calculated, and a personalized delivery time model of the delivery person is constructed;

[0087] When the adjustment strategy is route re-planning, a plurality of candidate routes are calculated according to the real-time traffic data, and the predicted delivery time of the delivery person on each candidate route is predicted in combination with the personalized delivery time model of the delivery person; the predicted delivery time is determined by the route length, the real-time traffic condition and the personalized delivery time model of the delivery person;

[0088] The route with the shortest predicted delivery time is selected as the optimal route.

[0089] In some embodiments, the strategy module 300 is used to perform the following when the adjustment strategy is route re-planning and the optimal route for the delivery person is re-determined according to the real-time traffic data:

[0090] The remaining power information of the delivery person's terminal is obtained;

[0091] When the adjustment strategy is route re-planning, a plurality of candidate routes are calculated according to the real-time traffic data, and the power required for the delivery person to reach all subsequent to-be-delivered points according to each candidate route is predicted in combination with the remaining power information of the delivery person's terminal;

[0092] Based on the predicted power and the remaining power of the delivery person's terminal, the optimal route for the delivery person is re-determined according to the real-time traffic data.

[0093] In some embodiments, the strategy module 300 is used to perform the following when the optimal route for the delivery person is re-determined according to the predicted power and the remaining power of the delivery person's terminal and the real-time traffic data:

[0094] The historical delivery data of the delivery person is obtained, the route segments frequently selected by the delivery person are extracted, and a personalized route preference library of the delivery person is constructed;

[0095] According to the delivery person personalized route preference library, the familiarity of the delivery person with each candidate route is determined, and the following steps A1-A4 are performed:

[0096] A1. If the familiarity of the delivery person with the candidate route is less than a preset familiarity threshold, and the predicted power is greater than or equal to the remaining power of the delivery person terminal, the route is eliminated;

[0097] A2. If the familiarity of the delivery person with the candidate route is less than the familiarity threshold, and the predicted power is less than the remaining power of the delivery person terminal, the route is retained;

[0098] A3. If the familiarity of the delivery person with the candidate route is greater than or equal to the familiarity threshold, and the predicted power is greater than or equal to the remaining power of the delivery person terminal, the predicted power is adjusted downward, and if the adjusted power is greater than or equal to the remaining power of the delivery person terminal, the route is eliminated; if the adjusted power is less than the remaining power of the delivery person terminal, the route is retained;

[0099] A4. If the familiarity of the delivery person with the candidate route is greater than or equal to the familiarity threshold, and the predicted power is less than the remaining power of the delivery person terminal, the route is retained;

[0100] Based on the remaining candidate routes, the optimal route for the delivery person is re-determined according to real-time traffic data.

[0101] In some embodiments, the strategy module 300 is configured to evaluate the feasibility of reassigning the orders in the affected delivery person's task that have not been picked up, and determine whether there is an alternative delivery person who can complete the orders on time when adjusting the strategy for order reassignment:

[0102] Extract key information of the orders in the affected delivery person's task that have not been picked up; the key information includes the estimated delivery time, the order value, and the distance information of each alternative delivery person to the pickup location;

[0103] According to the key information, calculate the reassignment score of each alternative delivery person; the shorter the estimated delivery time, the higher the order value, and the closer the distance, the higher the score;

[0104] Screen alternative delivery persons with a reassignment score higher than a preset score threshold, and if there is one, determine that there is an alternative delivery person who can complete the orders on time.

[0105] In some embodiments, the strategy module 300 is configured to calculate the reassignment score of each alternative delivery person according to the key information:

[0106] According to the key information, query a preset order type library to determine the delivery skills or equipment required by the order; the order type library stores information about the delivery skills or equipment required by different order types;

[0107] According to the historical delivery data of each substitute delivery person, it is analyzed whether each substitute delivery person has the delivery skills or equipment required by the order, and if not, the reassignment score of the substitute delivery person is zeroed.

[0108] In this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.

[0109] The above only describes the embodiments of the present application and is not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An order sorting and path optimization method based on real-time logistics data, characterized in that, The method comprises the following steps: obtaining real-time traffic data of a city road network, and determining congestion levels of road segments or areas based on the real-time traffic data; monitoring a delivery task in real time to obtain a position of a delivery staff, a task order list, an order state, and a planned route; when a congestion level of a road segment or an area on a current or planned route of the delivery staff changes, determining an adjustment strategy by consulting a preset congestion level classification response rule library, and when the adjustment strategy is route re-planning, re-determining an optimal route for the delivery staff based on real-time traffic data and in combination with a personalized route preference library of the delivery staff, a personalized delivery time model, or residual power information of a delivery staff terminal, and when the adjustment strategy is order re-assignment evaluation, evaluating re-assignment feasibility of an order that has not been picked up in a task of the affected delivery staff, and determining whether there is a substitute delivery staff who can complete the order on time; sending new route, order adjustment information, or re-assignment information to the delivery staff terminal in real time; when the adjustment strategy is order re-assignment evaluation, the step of evaluating re-assignment feasibility of an order that has not been picked up in a task of the affected delivery staff, and determining whether there is a substitute delivery staff who can complete the order on time, comprises: extracting key information of the order that has not been picked up in the task of the affected delivery staff; the key information includes an estimated delivery time, an order value, and distance information of each substitute delivery staff from a pickup location; calculating a re-assignment score of each substitute delivery staff according to the key information; the shorter the estimated delivery time, the higher the order value, and the closer the distance, the higher the score; selecting a substitute delivery staff whose re-assignment score is higher than a preset score threshold, and if there is one, determining that there is a substitute delivery staff who can complete the order on time; the step of calculating a re-assignment score of each substitute delivery staff according to the key information comprises: querying a preset order type library according to the key information to determine a required delivery skill or equipment of the order; the order type library stores delivery skill or equipment information required by different order types; analyzing whether each substitute delivery staff has the required delivery skill or equipment based on historical delivery data of each substitute delivery staff, and if not, setting the re-assignment score of the substitute delivery staff to zero.

2. The method for order sorting and path optimization based on real-time logistics data according to claim 1, characterized in that, when the adjustment strategy is route re-planning, the step of re-determining an optimal route for the delivery staff based on real-time traffic data comprises: obtaining historical delivery data of the delivery staff, and constructing a personalized route preference library of the delivery staff by extracting route segments frequently selected by the delivery staff; when the route is re-planned, calculating a plurality of candidate routes according to real-time traffic data, and combining the personalized route preference library of the delivery staff to score the candidate routes according to a degree of coincidence between a preferred route of the delivery staff and the candidate routes; the higher the degree of coincidence, the higher the score; selecting a route with the highest score as the optimal route.

3. The method for order sorting and path optimization based on real-time logistics data according to claim 1, characterized in that, when the adjustment strategy is route re-planning, the step of re-determining an optimal route for the delivery staff based on real-time traffic data comprises: obtaining historical delivery data of the delivery staff, calculating average delivery times of the delivery staff in different time periods and different areas, and constructing a personalized delivery time model of the delivery staff; When the adjustment strategy is route replanning, multiple candidate routes are calculated according to real-time traffic data, and the expected delivery time of the delivery personnel on each candidate route is predicted in combination with the personalized delivery time model of the delivery personnel; the expected delivery time is determined by the route length, real-time traffic conditions and the personalized delivery time model of the delivery personnel; The route with the shortest expected delivery time is selected as the optimal route.

4. The method for order sorting and path optimization based on real-time logistics data according to claim 1, characterized in that, When the adjustment strategy is route replanning, the steps of re-determining the optimal route for the delivery personnel according to real-time traffic data include: Obtaining the remaining power information of the delivery personnel terminal; When the adjustment strategy is route replanning, multiple candidate routes are calculated according to real-time traffic data, and the power required for the delivery personnel to reach all subsequent to-be-delivered points according to each candidate route is predicted in combination with the remaining power information of the delivery personnel terminal; Based on the predicted power and the remaining power of the delivery personnel terminal, the optimal route for the delivery personnel is re-determined according to real-time traffic data.

5. The method for order sorting and path optimization based on real-time logistics data according to claim 4, characterized in that, Based on the predicted power and the remaining power of the delivery personnel terminal, the steps of re-determining the optimal route for the delivery personnel according to real-time traffic data include: Obtaining historical delivery data of the delivery personnel, extracting frequently selected route segments of the delivery personnel, and constructing a personalized route preference library of the delivery personnel; According to the personalized route preference library of the delivery personnel, the familiarity of the delivery personnel with each candidate route is judged, and the following steps A1-A4 are specifically executed: A1. If the familiarity of the delivery personnel with the candidate route is less than a preset familiarity threshold, and the predicted power is greater than or equal to the remaining power of the delivery personnel terminal, the route is excluded; A2. If the familiarity of the delivery personnel with the candidate route is less than the familiarity threshold, and the predicted power is less than the remaining power of the delivery personnel terminal, the route is retained; A3. If the familiarity of the delivery personnel with the candidate route is greater than or equal to the familiarity threshold, and the predicted power is greater than or equal to the remaining power of the delivery personnel terminal, the predicted power is adjusted downward, and if the adjusted power is greater than or equal to the remaining power of the delivery personnel terminal, the route is excluded; if the adjusted power is less than the remaining power of the delivery personnel terminal, the route is retained; A4. If the familiarity of the delivery personnel with the candidate route is greater than or equal to the familiarity threshold, and the predicted power is less than the remaining power of the delivery personnel terminal, the route is retained; Based on the remaining candidate routes, the optimal route for the delivery personnel is re-determined according to real-time traffic data.

6. An order sorting and path optimization system based on real-time logistics data, characterized in that, It includes: A first obtaining module is configured to obtain real-time traffic data of a city road network, and determine the congestion level of a road section or an area based on the real-time traffic data; A second obtaining module is configured to monitor a delivery task in real time, and obtain the position of a delivery personnel, a task order list, an order state and an originally planned route; The strategy module is configured to determine an adjustment strategy by consulting a preset congestion level classification response rule library when a change occurs in the congestion level of a road segment or a region on a current or planned route of the delivery personnel, and to determine an optimal route for the delivery personnel again according to real-time traffic data and in combination with a personalized route preference library of the delivery personnel, a personalized delivery time model, or remaining power information of the delivery personnel terminal when the adjustment strategy is route replanning. The sending module is configured to send the new route, order adjustment information, or reassignment notification to the delivery personnel terminal in real time. The strategy module is configured to perform the following when the adjustment strategy is order reassignment evaluation: extract key information of an order that has not been picked up in the task of the affected delivery personnel; the key information includes an estimated delivery time, an order value, and distance information of each alternative delivery personnel from a pickup location; calculate a reassignment score of each alternative delivery personnel according to the key information; the shorter the estimated delivery time, the higher the order value, and the closer the distance, the higher the score; select an alternative delivery personnel whose reassignment score is higher than a preset score threshold, and if there is one, determine that there is an alternative delivery personnel who can complete the order on time. The strategy module is configured to perform the following when calculating a reassignment score of each alternative delivery personnel according to the key information: query a preset order type library according to the key information to determine the delivery skills or equipment required for the order; the order type library stores delivery skill or equipment information required for different order types; analyze whether each alternative delivery personnel has the delivery skills or equipment required for the order according to historical delivery data of each alternative delivery personnel, and if not, set the reassignment score of the alternative delivery personnel to zero.

7. The real-time logistics data based order sorting and path optimization system as claimed in claim 6, wherein, The strategy module is configured to perform the following when determining an optimal route for the delivery personnel again according to real-time traffic data when the adjustment strategy is route replanning: obtain historical delivery data of the delivery personnel, and construct a personalized route preference library of the delivery personnel by extracting route segments frequently selected by the delivery personnel; calculate a plurality of candidate routes according to real-time traffic data when the route is replanned, and combine the personalized route preference library of the delivery personnel to weight and score the candidate routes according to the degree of coincidence between the preferred route of the delivery personnel and the candidate routes; the higher the degree of coincidence with the preferred route of the delivery personnel, the higher the score; select the route with the highest score as the optimal route.

8. The real-time logistics data based order sorting and path optimization system as claimed in claim 6, wherein, The strategy module is configured to perform the following when determining an optimal route for the delivery personnel again according to real-time traffic data when the adjustment strategy is route replanning: obtain historical delivery data of the delivery personnel, and calculate average delivery times of the delivery personnel in different time periods and different regions to construct a personalized delivery time model of the delivery personnel; When the adjustment strategy is route replanning, multiple alternative routes are calculated according to real-time traffic data, and the expected delivery time of the delivery personnel on each alternative route is predicted in combination with the personalized delivery time model of the delivery personnel; the expected delivery time is determined by the route length, real-time traffic conditions and the personalized delivery time model of the delivery personnel; The route with the shortest expected delivery time is selected as the optimal route.

Images